LLM Agent Research Scientist

Responsibilities

• Lead research initiatives for LLM-powered agents aligned with product goals
• Develop novel agent capabilities through post-training, RL, reward modeling, and synthetic data
• Design agent architectures, reasoning methods, and multi-agent frameworks
• Build agent harnesses for executing complex, long-horizon tasks in production
• Create rigorous benchmarks and evaluation methods for agent correctness and safety
• Partner with engineering and product teams to integrate research into Canva products

Requirements

• Hands-on experience with LLM-based agents or foundation models
• Expertise in post-training, RL, reward modeling, or synthetic data
• Strong academic track record with publications in top-tier venues (NeurIPS, ICML, ICLR, etc.) or impactful open-source contributions
• Proficiency in Python and PyTorch
• Familiarity with Transformers, DeepSpeed, Megatron, and RL frameworks
• Experience with cloud computing platforms for large-scale training

About the Company

At Canva, we're building a future powered by AI that's as magical as it is impactful. Our CORE team within the Generative AI supergroup invents foundational technologies to power the future of AI-assisted design.